Hands-on Exercise 3: Programming Interactive Data Visualisation with R and Programming Animated Statistical Graphics with R

Author

Rizqi Febriansyah

Published

January 23, 2023

Modified

January 26, 2024

Part One: Programming Interactive Data Visualisation with R

3.1 Learning Outcome

In this hands-on exercise, we will learn how to create interactive data visualisation by using functions provided by ggiraph and plotlyr packages.

3.2 Getting Started

First, we will write a code chunk to check, install and launch the following R packages:

  • ggiraph for making ‘ggplot’ graphics interactive.
  • plotly, R library for plotting interactive statistical graphs. - DT provides an R interface to the JavaScript library
  • DataTables that create interactive table on html page.
  • tidyverse, a family of modern R packages specially designed to support data science, analysis and communication task including creating static statistical graphs.
  • patchwork for combining multiple ggplot2 graphs into one figure.

The code chunk below will be used to accomplish the task.

pacman::p_load(ggiraph, plotly, 
               patchwork, DT, tidyverse) 

3.3 Importing Data

In this section, Exam_data.csv provided will be used. Using read_csv() of readr package, import Exam_data.csv into R.

The code chunk below read_csv() of readr package is used to import Exam_data.csv data file into R and save it as an tibble data frame called exam_data.

exam_data <- read_csv("data/Exam_data.csv")

3.4 Interactive Data Visualisation - ggiraph methods

ggiraph is an htmlwidget and a ggplot2 extension. It allows ggplot graphics to be interactive.

Interactive is made with ggplot geometries (images/ggiraph.png) that can understand three arguments:

  • Tooltip: a column of data-sets that contain tooltips to be displayed when the mouse is over elements.
  • Onclick: a column of data-sets that contain a JavaScript function to be executed when elements are clicked.
  • Data_id: a column of data-sets that contain an id to be associated with elements.

If it used within a shiny application, elements associated with an id (data_id) can be selected and manipulated on client and server sides. Refer to this article for more detail explanation.

3.4.1 Tooltip effect with tooltip aesthetic

Below shows a typical code chunk to plot an interactive statistical graph by using ggiraph package. Notice that the code chunk consists of two parts. First, an ggplot object will be created. Next, girafe() of ggiraph will be used to create an interactive svg object.

p <- ggplot(data=exam_data, 
       aes(x = MATHS)) +
  geom_dotplot_interactive(
    aes(tooltip = ID),
    stackgroups = TRUE, 
    binwidth = 1, 
    method = "histodot") +
  scale_y_continuous(NULL, 
                     breaks = NULL)
girafe(
  ggobj = p,
  width_svg = 6,
  height_svg = 6*0.618
)

Notice that two steps are involved. First, an interactive version of ggplot2 geom (i.e. geom_dotplot_interactive()) will be used to create the basic graph. Then, girafe() will be used to generate an svg object to be displayed on an html page.

3.5 Interactivity part 1

By hovering the mouse pointer on an data point of interest, the student’s ID will be displayed.

3.5.1 Displaying multiple information on tooltip

The content of the tooltip can be customised by including a list object as shown in the code chunk below.

exam_data$tooltip <- c(paste0(     
  "Name = ", exam_data$ID,         
  "\n Class = ", exam_data$CLASS)) 

p <- ggplot(data=exam_data, 
       aes(x = MATHS)) +
  geom_dotplot_interactive(
    aes(tooltip = exam_data$tooltip), 
    stackgroups = TRUE,
    binwidth = 1,
    method = "histodot") +
  scale_y_continuous(NULL,               
                     breaks = NULL)
girafe(
  ggobj = p,
  width_svg = 8,
  height_svg = 8*0.618
)

The first three lines of codes in the code chunk create a new field called tooltip. At the same time, it populates text in ID and CLASS fields into the newly created field. Next, this newly created field is used as tooltip field as shown in the code of line 7.

3.6 Interactivity part 2

By hovering the mouse pointer on an data point of interest, the student’s ID will be displayed.

3.6.1 Customising Tooltip style

Code chunk below uses opts_tooltip() of ggiraph to customize tooltip rendering by add css declarations.

tooltip_css <- "background-color:white; #<<
font-style:bold; color:black;" #<<

p <- ggplot(data=exam_data, 
       aes(x = MATHS)) +
  geom_dotplot_interactive(              
    aes(tooltip = ID),                   
    stackgroups = TRUE,                  
    binwidth = 1,                        
    method = "histodot") +               
  scale_y_continuous(NULL,               
                     breaks = NULL)
girafe(                                  
  ggobj = p,                             
  width_svg = 6,                         
  height_svg = 6*0.618,
  options = list(    #<<
    opts_tooltip(    #<<
      css = tooltip_css)) #<<
)                                        

Notice that the background colour of the tooltip is black and the font colour is white and bold.

  • Refer to Customising girafe objects to learn more about how to customise ggiraph objects.

3.6.2 Displaying statistics on tooltip

Code chunk below shows an advanced way to customise tooltip. In this example, a function is used to compute 90% confident interval of the mean. The derived statistics are then displayed in the tooltip.

tooltip <- function(y, ymax, accuracy = .01) {
  mean <- scales::number(y, accuracy = accuracy)
  sem <- scales::number(ymax - y, accuracy = accuracy)
  paste("Mean maths scores:", mean, "+/-", sem)
}

gg_point <- ggplot(data=exam_data, 
                   aes(x = RACE),
) +
  stat_summary(aes(y = MATHS, 
                   tooltip = after_stat(  
                     tooltip(y, ymax))),  
    fun.data = "mean_se", 
    geom = GeomInteractiveCol,  
    fill = "light blue"
  ) +
  stat_summary(aes(y = MATHS),
    fun.data = mean_se,
    geom = "errorbar", width = 0.2, size = 0.2
  )

girafe(ggobj = gg_point,
       width_svg = 8,
       height_svg = 8*0.618)

3.6.3 Hover effect with data_id aesthetic

Code chunk below shows the second interactive feature of ggiraph, namely data_id.

p <- ggplot(data=exam_data, 
       aes(x = MATHS)) +
  geom_dotplot_interactive(           
    aes(data_id = CLASS),             
    stackgroups = TRUE,               
    binwidth = 1,                        
    method = "histodot") +               
  scale_y_continuous(NULL,               
                     breaks = NULL)
girafe(                                  
  ggobj = p,                             
  width_svg = 6,                         
  height_svg = 6*0.618                      
)

Interactivity: Elements associated with a data_id (i.e CLASS) will be highlighted upon mouse over.

Note that the default value of the hover css is hover_css = “fill:orange;”.

3.6.4 Styling hover effect

In the code chunk below, css codes are used to change the highlighting effect.

Interactivity: Elements associated with a data_id (i.e CLASS) will be highlighted upon mouse over.

p <- ggplot(data=exam_data, 
       aes(x = MATHS)) +
  geom_dotplot_interactive(              
    aes(data_id = CLASS),              
    stackgroups = TRUE,                  
    binwidth = 1,                        
    method = "histodot") +               
  scale_y_continuous(NULL,               
                     breaks = NULL)
girafe(                                  
  ggobj = p,                             
  width_svg = 6,                         
  height_svg = 6*0.618,
  options = list(                        
    opts_hover(css = "fill: #202020;"),  
    opts_hover_inv(css = "opacity:0.2;") 
  )                                        
)                                        

Note: Different from previous example, in this example the ccs customisation request are encoded directly.

3.6.5 Combining tooltip and hover effect

There are times that we want to combine tooltip and hover effect on the interactive statistical graph as shown in the code chunk below.

Interactivity: Elements associated with a data_id (i.e CLASS) will be highlighted upon mouse over. At the same time, the tooltip will show the CLASS.

p <- ggplot(data=exam_data, 
       aes(x = MATHS)) +
  geom_dotplot_interactive(              
    aes(tooltip = CLASS, 
        data_id = CLASS),              
    stackgroups = TRUE,                  
    binwidth = 1,                        
    method = "histodot") +               
  scale_y_continuous(NULL,               
                     breaks = NULL)
girafe(                                  
  ggobj = p,                             
  width_svg = 6,                         
  height_svg = 6*0.618,
  options = list(                        
    opts_hover(css = "fill: #202020;"),  
    opts_hover_inv(css = "opacity:0.2;") 
  )                                        
)                                        

3.6.6 Click effect with onclick

onclick argument of ggiraph provides hotlink interactivity on the web.

The code chunk below shown an example of onclick.

exam_data$onclick <- sprintf("window.open(\"%s%s\")",
"https://www.moe.gov.sg/schoolfinder?journey=Primary%20school",
as.character(exam_data$ID))

p <- ggplot(data=exam_data, 
       aes(x = MATHS)) +
  geom_dotplot_interactive(              
    aes(onclick = onclick),              
    stackgroups = TRUE,                  
    binwidth = 1,                        
    method = "histodot") +               
  scale_y_continuous(NULL,               
                     breaks = NULL)
girafe(                                  
  ggobj = p,                             
  width_svg = 6,                         
  height_svg = 6*0.618)     

Interactivity: Web document link with a data object will be displayed on the web browser upon mouse click.

Warning

Note that click actions must be a string column in the dataset containing valid javascript instructions.

3.6.7 Coordinated Multiple Views with ggiraph

Coordinated multiple views methods has been implemented in the data visualisation below.

Notice that when a data point of one of the dotplot is selected, the corresponding data point ID on the second data visualisation will be highlighted too.

In order to build a coordinated multiple views as shown in the example above, the following programming strategy will be used:

  1. Appropriate interactive functions of ggiraph will be used to create the multiple views.
  2. patchwork function of patchwork package will be used inside girafe function to create the interactive coordinated multiple views.
p1 <- ggplot(data=exam_data, 
       aes(x = MATHS)) +
  geom_dotplot_interactive(              
    aes(data_id = ID),              
    stackgroups = TRUE,                  
    binwidth = 1,                        
    method = "histodot") +  
  coord_cartesian(xlim=c(0,100)) + 
  scale_y_continuous(NULL,               
                     breaks = NULL)

p2 <- ggplot(data=exam_data, 
       aes(x = ENGLISH)) +
  geom_dotplot_interactive(              
    aes(data_id = ID),              
    stackgroups = TRUE,                  
    binwidth = 1,                        
    method = "histodot") + 
  coord_cartesian(xlim=c(0,100)) + 
  scale_y_continuous(NULL,               
                     breaks = NULL)

girafe(code = print(p1 + p2), 
       width_svg = 6,
       height_svg = 3,
       options = list(
         opts_hover(css = "fill: #202020;"),
         opts_hover_inv(css = "opacity:0.2;")
         )
       ) 

The data_id aesthetic is critical to link observations between plots and the tooltip aesthetic is optional but nice to have when mouse over a point.

3.7 Interactive Data Visualisation - plotly methods!

Plotly’s R graphing library create interactive web graphics from ggplot2 graphs and/or a custom interface to the (MIT-licensed) JavaScript library plotly.js inspired by the grammar of graphics. Different from other plotly platform, plot.R is free and open source.

There are two ways to create interactive graph by using plotly, they are:

  • by using plot_ly(), and
  • by using ggplotly()

3.7.1 Creating an interactive scatter plot: plot_ly() method

The tabset below shows an example a basic interactive plot created by using plot_ly().

Interactive:

Click on the colour symbol at the legend.

plot_ly(data = exam_data, 
             x = ~MATHS, 
             y = ~ENGLISH)

3.7.2 Working with visual variable: plot_ly() method

In the code chunk below, color argument is mapped to a qualitative visual variable (i.e. RACE).

plot_ly(data = exam_data, 
        x = ~ENGLISH, 
        y = ~MATHS, 
        color = ~RACE)

3.7.3 Creating an interactive scatter plot: ggplotly() method

The code chunk below plots an interactive scatter plot by using ggplotly().

p <- ggplot(data=exam_data, 
            aes(x = MATHS,
                y = ENGLISH)) +
  geom_point(size=1) +
  coord_cartesian(xlim=c(0,100),
                  ylim=c(0,100))
ggplotly(p)

Notice that the only extra line you need to include in the code chunk is ggplotly().

3.7.4 Coordinated Multiple Views with plotly

The creation of a coordinated linked plot by using plotly involves three steps:

  • highlight_key() of plotly package is used as shared data.
  • two scatterplots will be created by using ggplot2 functions.
  • lastly, subplot() of plotly package is used to place them next to each other side-by-side.

Click on a data point of one of the scatterplot and see how the corresponding point on the other scatterplot is selected.

d <- highlight_key(exam_data)
p1 <- ggplot(data=d, 
            aes(x = MATHS,
                y = ENGLISH)) +
  geom_point(size=1) +
  coord_cartesian(xlim=c(0,100),
                  ylim=c(0,100))

p2 <- ggplot(data=d, 
            aes(x = MATHS,
                y = SCIENCE)) +
  geom_point(size=1) +
  coord_cartesian(xlim=c(0,100),
                  ylim=c(0,100))
subplot(ggplotly(p1),
        ggplotly(p2))

Thing to learn from the code chunk:

3.8 Interactive Data Visualisation - crosstalk methods!

Crosstalk is an add-on to the htmlwidgets package. It extends htmlwidgets with a set of classes, functions, and conventions for implementing cross-widget interactions (currently, linked brushing and filtering).

3.8.1 Interactive Data Table: DT package

  • A wrapper of the JavaScript Library DataTables

  • Data objects in R can be rendered as HTML tables using the JavaScript library ‘DataTables’ (typically via R Markdown or Shiny).

DT::datatable(exam_data, class= "compact")

3.8.2 Linked brushing: crosstalk method

Code chunk below is used to implement the coordinated brushing shown above.

d <- highlight_key(exam_data) 
p <- ggplot(d, 
            aes(ENGLISH, 
                MATHS)) + 
  geom_point(size=1) +
  coord_cartesian(xlim=c(0,100),
                  ylim=c(0,100))

gg <- highlight(ggplotly(p),        
                "plotly_selected")  

crosstalk::bscols(gg,               
                  DT::datatable(d), 
                  widths = 5)        

Things to learn from the code chunk:

  • highlight() is a function of plotly package. It sets a variety of options for brushing (i.e., highlighting) multiple plots. These options are primarily designed for linking multiple plotly graphs, and may not behave as expected when linking plotly to another htmlwidget package via crosstalk. In some cases, other htmlwidgets will respect these options, such as persistent selection in leaflet.

  • bscols() is a helper function of crosstalk package. It makes it easy to put HTML elements side by side. It can be called directly from the console but is especially designed to work in an R Markdown document. Warning: This will bring in all of Bootstrap!.

3.9 Reference

3.9.1 ggiraph

This link provides online version of the reference guide and several useful articles. Use this link to download the pdf version of the reference guide.

3.9.2 plotly for R

Part Two: Programming Animated Statistical Graphics with R

4.1 Overview

In narrating a data story driven by visuals, animated graphics often captivate the audience’s attention and leave a more profound impression compared to static visuals. In this interactive session, we will gain proficiency in crafting animated data visualisations using the gganimate and plotlyr R packages. Simultaneously, we will also acquire the skills to: (i) reshape data through the tidyr package and (ii) process, manipulate, and transform data using the dplyr package.

4.1.1 Basic concepts of animation

In the process of crafting animations, the plot itself remains static. Instead, numerous individual plots are constructed and subsequently assembled into sequences resembling movie frames, akin to the traditional method used in flip books or cartoons. Each frame represents a distinct plot capturing motion, constructed from a pertinent subset of the aggregated data. The interconnected frames, when combined, create the animated flow driven by the selected data subsets.

4.1.2 Terminology

Prior to delving into the procedures for generating an animated statistical graph, it is crucial to familiarise ourselves with key concepts and terminology associated with this form of visualisation.

  • Frame: Within an animated line graph, each frame corresponds to a distinct time point or category. As the frame transitions, the data points on the graph dynamically adjust to reflect the updated data.

  • Animation Attributes: These attributes encompass the settings dictating the behavior of the animation. For instance, one can specify the duration of each frame, the easing function facilitating transitions between frames, and whether the animation should commence from the current frame or from the outset.

Tip

Prior to embarking on the creation of animated graphs, it is essential to consider whether the effort is justified. For exploratory data analysis, investing time in animated graphics might not be warranted. However, in a presentation setting, strategically incorporating a few well-designed animated graphics can significantly enhance the audience’s connection with your topic compared to static alternatives.

4.2 Getting Started

4.2.1 Loading the R packages

First, write a code chunk to check, install and load the following R packages:

  • plotly, R library for plotting interactive statistical graphs.
  • gganimate, an ggplot extension for creating animated statistical graphs.
  • gifski converts video frames to GIF animations using pngquant’s fancy features for efficient cross-frame palettes and temporal dithering. It produces animated GIFs that use thousands of colors per frame.
  • gapminder: An excerpt of the data available at Gapminder.org. We just want to use its country_colors scheme.
  • tidyverse, a family of modern R packages specially designed to support data science, analysis and communication task including creating static statistical graphs.
pacman::p_load(readxl, gifski, gapminder,
               plotly, gganimate, tidyverse)

package ‘gifski’ successfully unpacked and MD5 sums checked

The downloaded binary packages are in C:4R_packages package ‘gapminder’ successfully unpacked and MD5 sums checked

The downloaded binary packages are in C:4R_packages package ‘gganimate’ successfully unpacked and MD5 sums checked

The downloaded binary packages are in C:4R_packages

4.2.2 Importing the data

In this hands-on exercise, the Data worksheet from GlobalPopulation Excel workbook will be used.

Write a code chunk to import Data worksheet from GlobalPopulation Excel workbook by using appropriate R package from tidyverse family.

col <- c("Country", "Continent")
globalPop <- read_xls("data/GlobalPopulation.xls",
                      sheet="Data") %>%
  mutate_each_(funs(factor(.)), col) %>%
  mutate(Year = as.integer(Year))
Things to learn from the code above
  • read_xls() of readxl package is used to import the Excel worksheet.
  • mutate_each_() of dplyr package is used to convert all character data type into factor.
  • mutate` of dplyr package is used to convert data values of Year field into integer.

Unfortunately, mutate_each_() was deprecated in dplyr 0.7.0. and funs() was deprecated in dplyr 0.8.0. In view of this, we will re-write the code by using mutate_at() as shown in the code chunk below.

col <- c("Country", "Continent")
globalPop <- read_xls("data/GlobalPopulation.xls",
                      sheet="Data") %>%
  mutate_at(col, as.factor) %>%
  mutate(Year = as.integer(Year))

Instead of using mutate_at(), across() can be used to derive the same outputs.

col <- c("Country", "Continent")
globalPop <- read_xls("data/GlobalPopulation.xls",
                      sheet="Data") %>%
  mutate(across(col, as.factor)) %>%
  mutate(Year = as.integer(Year))

4.3 Animated Data Visualisation: gganimate methods

gganimate extends the grammar of graphics as implemented by ggplot2 to include the description of animation. It does this by providing a range of new grammar classes that can be added to the plot object in order to customise how it should change with time.

  • transition_*() defines how the data should be spread out and how it relates to itself across time.
  • view_*() defines how the positional scales should change along the animation.
  • shadow_*() defines how data from other points in time should be presented in the given point in time.
  • enter_*()/exit_*() defines how new data should appear and how old data should disappear during the course of the animation.
  • ease_aes() defines how different aesthetics should be eased during transitions.

4.3.1 Building a static population bubble plot

In the code chunk below, the basic ggplot2 functions are used to create a static bubble plot.

ggplot(globalPop, aes(x = Old, y = Young, 
                      size = Population, 
                      colour = Country)) +
  geom_point(alpha = 0.7, 
             show.legend = FALSE) +
  scale_colour_manual(values = country_colors) +
  scale_size(range = c(2, 12)) +
  labs(title = 'Year: {frame_time}', 
       x = '% Aged', 
       y = '% Young') 

4.3.2 Building the animated bubble plot

In the code chunk below,

  • transition_time() of gganimate is used to create transition through distinct states in time (i.e. Year).
  • ease_aes() is used to control easing of aesthetics. The default is linear. Other methods are: quadratic, cubic, quartic, quintic, sine, circular, exponential, elastic, back, and bounce.
ggplot(globalPop, aes(x = Old, y = Young, 
                      size = Population, 
                      colour = Country)) +
  geom_point(alpha = 0.7, 
             show.legend = FALSE) +
  scale_colour_manual(values = country_colors) +
  scale_size(range = c(2, 12)) +
  labs(title = 'Year: {frame_time}', 
       x = '% Aged', 
       y = '% Young') +
  transition_time(Year) +       
  ease_aes('linear')          

4.4 Animated Data Visualisation: plotly

In Plotly R package, both ggplotly() and plot_ly() support key frame animations through the frame argument/aesthetic. They also support an ids argument/aesthetic to ensure smooth transitions between objects with the same id (which helps facilitate object constancy).

4.4.1 Building an animated bubble plot: ggplotly() method

In this sub-section, you will learn how to create an animated bubble plot by using ggplotly() method.

The animated bubble plot above includes a play/pause button and a slider component for controlling the animation

gg <- ggplot(globalPop, 
       aes(x = Old, 
           y = Young, 
           size = Population, 
           colour = Country)) +
  geom_point(aes(size = Population,
                 frame = Year),
             alpha = 0.7, 
             show.legend = FALSE) +
  scale_colour_manual(values = country_colors) +
  scale_size(range = c(2, 12)) +
  labs(x = '% Aged', 
       y = '% Young')

ggplotly(gg)
Things to learn from code above
  • Appropriate ggplot2 functions are used to create a static bubble plot. The output is then saved as an R object called gg.
  • ggplotly() is then used to convert the R graphic object into an animated svg object.

Notice that although show.legend = FALSE argument was used, the legend still appears on the plot. To overcome this problem, theme(legend.position='none') should be used as shown in the plot and code chunk below.

gg <- ggplot(globalPop, 
       aes(x = Old, 
           y = Young, 
           size = Population, 
           colour = Country)) +
  geom_point(aes(size = Population,
                 frame = Year),
             alpha = 0.7) +
  scale_colour_manual(values = country_colors) +
  scale_size(range = c(2, 12)) +
  labs(x = '% Aged', 
       y = '% Young') + 
  theme(legend.position='none')

ggplotly(gg)

4.4.2 Building an animated bubble plot: plot_ly() method

In this sub-section, you will learn how to create an animated bubble plot by using plot_ly() method.

bp <- globalPop %>%
  plot_ly(x = ~Old, 
          y = ~Young, 
          size = ~Population, 
          color = ~Continent,
          sizes = c(2, 100),
          frame = ~Year, 
          text = ~Country, 
          hoverinfo = "text",
          type = 'scatter',
          mode = 'markers'
          ) %>%
  layout(showlegend = FALSE)
bp

4.5 Reference